LECTURE 02
FROM COGNITIVE MODELS TO CONSTRUCTING CAREFULLY-DESIGNED EXPERIMENTS FEATURES OF HIGH-VALUE COGNITIVE TESTS
FROM COGNITIVE TASKS TO COGNITIVE NEUROSCIENCE
PSYB57: INTRODUCTION TO COGNITIVE PSYCHOLOGY | UNIVERSITY OF TORONTO, SCARBOROUGH | PROF. MICHAEL SOUZA
Copyright By PowCoder代写 加微信 powcoder
Lecture objectives
To understand and describe some key principles that inform the development of cognitive models;
To describe conceptual and operational definitions of several cognitive processes of interest so that we may design paradigms to study them;
To understand the purpose and value of visualizing an experimental paradigm, and to position yourself to create such visualizations;
To critically analyze how selection of various features of an experimental paradigm may influence participant performance, and the underlying cognitive processes we wish to study;
To identify and describe some key features that must be considered for a cognitive test (e.g., validity) to be used more widely;
To briefly consider and understand how common cognitive neuroscience tools (e.g., EEG, fMRI) may be used in parallel with cognitive paradigms to study their underlying neural mechanisms.
A framework for studying cognition
Stimulus/ Response/ input output
Carefully- designed experiment
Mental processes involved in the task
Behavioral dependent variables
Articulating cognitive models (1)
Cognitive models are created to describe the various steps hypothesized to be involved in a cognitive process (e.g., reading)
When creating such models, researchers should be mindful of the following:
What is this model trying to describe? (clearly specified)
Are the steps and linkages in this model reasonable to hypothesize? (realistic elements) Is this the simplest model we could create to describe this process? (parsimonious)
Is this model testable and can I provide evidence for or against it? (falsifiable)
Does our knowledge of the brain corroborate this model? (link to neuroscience)
Atkinson & Shiffrin (1968) model of memory (a.k.a.: the modal model of memory)
Articulating cognitive models (2)
Baddeley & Hitch’s (1974) model of working memory
Baddeley, A. (2003), Nature Reviews Neuroscience, 4(10), 829.
Articulating cognitive models (3)
What causes utilization behavior?
Norman & Shallice’s (1980) model of the Supervisory Attentional System
Carefully-designed experiment #1: Visualizing an experimental paradigm
Suppose I’m a researcher interested in working memory (WM) loads Articulating a clear conceptual definition, and then operational definition
Why is it important to study something with many trials, and many people?
(pay attention now)
(“XOQB, XOQB…”)
(“XOQB, XOQB…”)
PROBE (TEST)
(YES: L button; NO: R button)
GET READY!
(the “flow” of the experimental paradigm)
Ext. focused attention
Ext. focused attention Verbal encoding
(the “flow” of the experimental paradigm)
GET READY!
Carefully-designed experiment #1: Putative mental processes involved in the task (load of 4)
Int. focused attention WM maintenance
Ext. focused attention WM maintenance Comparator process Decision process Generate response
Carefully-designed experiment #1: Manipulating WM demands (loads of 4, 8)
STIMULUS DELAY PROBE
GET READY!
GET READY!
(the “flow” of the experimental paradigm)
Carefully-designed experiment #1: Putative mental effects of manipulating WM load?
GET READY!
GET READY!
Ext. focused attention
Ext. focused attention Verbal encoding
Int. focused attention WM maintenance
Ext. focused attention WM maintenance Comparator process Decision process Generate response
Higher WM demands Lower WM demands
Higher WM demands Lower WM demands
Carefully-designed experiment #1: Behavioral dependent variables
Overall assessment of task performance Understanding correct, incorrect trials
Reaction time (RT)
Time it takes to generate a response Discarding of incorrect trials
Other considerations: what if…?
Both accuracies at 100%?
Accuracy comparable (i.e., ~90%), but RTs significantly different?
Carefully-designed experiment #1: The relationship between accuracy and RT
Interpreting the speed-accuracy trade-off Effect of task demands on the participant’s behavior
Carefully-designed experiment #1: Manipulating WM content while keeping load constant
Lettuce Chicken Corn Chocolate
GET READY!
Carefully-designed experiment #1: Variations of this WM task
GET READY!
Loads more on the comparator process (order matters)
Now requires manipulating the information, and then maintaining it
(forward order)
(backwards order)
Experiment #1 reflection
Details matter with respect to experimental design
Be thoughtful about what you’re asking the participant to do, and why
Things as ‘basic’ as the stimuli used (i.e., words vs. pictures) may have an important effect
The challenge of understanding your data
The pattern of results in your data could seem straightforward…or not so much
We evaluate competing alternatives and to consider ways to eliminate such alternatives through thoughtful follow-ups (i.e., avoid confirmation bias and ‘tunnel vision’ science)
Lettuce Chicken Corn Chocolate
Carefully-designed experiment #2: Practice with experimental paradigms
Suppose I’m a cognitive researcher interested in response inhibition Articulating a clear conceptual definition, and then operational definition
Continuous Performance Task (CPT)
Task: press the button for every letter except “C” and “M” Effect of pressing the button the majority of the time
YES! YES! YES
500 ms 500 ms 500 ms
GET READY!
(the “flow” of the experimental paradigm)
Rule maintenance
Rule maintenance Comparator process Decide “yes” or “no”
(press button)
(don’t press button)
Press button, or inhibit button press
Button pressed
Button not pressed
Button pressed
Button not pressed
Classifying the outcome
ERROR OF OMMISSION
ERROR OF COMMISSION
CORRECT REJECTION
Carefully-designed experiment #2: Mental processes involved and response possibilities
GET READY!
Continuous Performance Task (CPT)
Press the button for every letter except “C” and “M”
Carefully-designed experiment #2: Behavioral dependent variables of interest
Analyzing hits and correction rejection rate Analyzing errors of omission, commission rate
Reaction time (RT)
Analyzing hits vs. errors of commission (recall: no responses for rejections, omissions)
Effect of speed-accuracy trade-off in this task?
Carefully-designed experiment #3: Practicing experimental design
Suppose I’m a cognitive researcher interested in choosing a less frequent behavior in the face of a much more frequent behavior
Articulating a clear conceptual definition, and then operational definition
Stroop Task
Task: Press one of four buttons to indicate the color of the ink (not the word) Button 1: Red | Button 2: Blue | Button 3: Green | Button 4: Orange
EXPERIMENTAL
RED BLUE GREEN GREEN
500 ms 500 ms 500 ms 500 ms
(the “flow” of the experimental paradigm)
GET READY!
An important skill to practice for this course! For more examples to play with, consider visiting: http://cognitivefun.net/
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com